', re.IGNORECASE), lambda match: '
'),
- # Replace
with
- (re.compile(r'\s*', re.IGNORECASE), lambda match: ''),
-
- # Remove hyphenation
- (re.compile(r'-\n\r?'), lambda match: ''),
+ (re.compile(r'', re.IGNORECASE), lambda match: '
'),
# Remove gray background
(re.compile(r']+>'), lambda match : ''),
# Detect Chapters to match default XPATH in GUI
- (re.compile(r'(?=<(/?br|p))(<(/?br|p)[^>]*)?>\s*(?P(<(i|b)>(<(i|b)>)?)?(.?Chapter|Epilogue|Prologue|Book|Part)\s*([\d\w-]+(\s\w+)?)?(()?)?)]*>\s*(?P(<(i|b)>)?\s*\w+(\s*\w+)?\s*(</(i|b)>)?\s*(</?(br|p)[^>]*>))?', re.IGNORECASE), chap_head),
- (re.compile(r'(?=<(/?br|p))(<(/?br|p)[^>]*)?>\s*(?P<chap>([A-Z \'"!]{5,})\s*(\d+|\w+)?)(</?p[^>]*>|<br[^>]*>)\n?((?=(<i>)?\s*\w+(\s+\w+)?(</i>)?(<br[^>]*>|</?p[^>]*>))((?P<title>.*)(<br[^>]*>|</?p[^>]*>)))?'), chap_head),
+ (re.compile(r'<br>\s*(?P<chap>(<[ibu]>){0,2}\s*.?(Introduction|Chapter|Epilogue|Prologue|Book|Part|Dedication|Volume|Preface|Acknowledgments)\s*([\d\w-]+\s*){0,3}\s*(</[ibu]>){0,2})\s*(<br>\s*){1,3}\s*(?P<title>(<[ibu]>){0,2}(\s*\w+){1,4}\s*(</[ibu]>){0,2}\s*<br>)?', re.IGNORECASE), chap_head),
+ # Cover the case where every letter in a chapter title is separated by a space
+ (re.compile(r'<br>\s*(?P<chap>([A-Z]\s+){4,}\s*([\d\w-]+\s*){0,3}\s*)\s*(<br>\s*){1,3}\s*(?P<title>(<[ibu]>){0,2}(\s*\w+){1,4}\s*(</[ibu]>){0,2}\s*(<br>))?'), chap_head),
# Have paragraphs show better
(re.compile(r'<br.*?>'), lambda match : '<p>'),
# Clean up spaces
(re.compile(u'(?<=[\.,;\?!”"\'])[\s^ ]*(?=<)'), lambda match: ' '),
- # Connect paragraphs split by -
- (re.compile(u'(?<=[^\s][-–])[\s]*(</p>)*[\s]*(<p>)*\s*(?=[^\s])'), lambda match: ''),
# Add space before and after italics
(re.compile(u'(?<!“)<i>'), lambda match: ' <i>'),
(re.compile(r'</i>(?=\w)'), lambda match: '</i> '),
@@ -328,12 +329,29 @@ class HTMLPreProcessor(object):
print 'Failed to parse remove_footer regexp'
traceback.print_exc()
+ # unwrap hyphenation - moved here so it's executed after header/footer removal
+ if is_pdftohtml:
+ # unwrap visible dashes and hyphens - don't delete they are often hyphens for
+ # for compound words, formatting, etc
+ end_rules.append((re.compile(u'(?<=[-–—])\s*<p>\s*(?=[[a-z\d])'), lambda match: ''))
+ # unwrap/delete soft hyphens
+ end_rules.append((re.compile(u'[](\s*<p>)+\s*(?=[[a-z\d])'), lambda match: ''))
+ # unwrap/delete soft hyphens with formatting
+ end_rules.append((re.compile(u'[]\s*(</(i|u|b)>)+(\s*<p>)+\s*(<(i|u|b)>)+\s*(?=[[a-z\d])'), lambda match: ''))
+
+ # Make the more aggressive chapter marking regex optional with the preprocess option to
+ # reduce false positives and move after header/footer removal
+ if getattr(self.extra_opts, 'preprocess_html', None):
+ if is_pdftohtml:
+ end_rules.append((re.compile(r'<p>\s*(?P<chap>(<[ibu]>){0,2}\s*([A-Z \'"!]{3,})\s*([\dA-Z:]+\s){0,4}\s*(</[ibu]>){0,2})\s*<p>\s*(?P<title>(<[ibu]>){0,2}(\s*\w+){1,4}\s*(</[ibu]>){0,2}\s*<p>)?'), chap_head),)
+
if getattr(self.extra_opts, 'unwrap_factor', 0.0) > 0.01:
length = line_length('pdf', html, getattr(self.extra_opts, 'unwrap_factor'))
if length:
+ # print "The pdf line length returned is " + str(length)
end_rules.append(
# Un wrap using punctuation
- (re.compile(r'(?<=.{%i}[a-z\.,;:)\-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>)\s*(?=(<(i|b|u)>)?\s*[\w\d(])' % length, re.UNICODE), wrap_lines),
+ (re.compile(r'(?<=.{%i}[a-z,;:)\-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>\s*)+\s*(?=(<(i|b|u)>)?\s*[\w\d$(])' % length, re.UNICODE), wrap_lines),
)
for rule in self.PREPROCESS + start_rules:
@@ -383,5 +401,14 @@ class HTMLPreProcessor(object):
if self.plugin_preprocess:
html = self.input_plugin_preprocess(html)
+ if getattr(self.extra_opts, 'smarten_punctuation', False):
+ html = self.smarten_punctuation(html)
+
return html
+ def smarten_punctuation(self, html):
+ from calibre.utils.smartypants import smartyPants
+ from calibre.ebooks.chardet import substitute_entites
+ html = smartyPants(html)
+ return substitute_entites(html)
+
diff --git a/src/calibre/ebooks/conversion/utils.py b/src/calibre/ebooks/conversion/utils.py
new file mode 100644
index 0000000000..5301f70a16
--- /dev/null
+++ b/src/calibre/ebooks/conversion/utils.py
@@ -0,0 +1,173 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+
+__license__ = 'GPL v3'
+__copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+import re
+from calibre.ebooks.conversion.preprocess import line_length
+from calibre.utils.logging import default_log
+
+class PreProcessor(object):
+
+ def __init__(self, log=None):
+ self.log = default_log if log is None else log
+ self.html_preprocess_sections = 0
+ self.found_indents = 0
+
+ def chapter_head(self, match):
+ chap = match.group('chap')
+ title = match.group('title')
+ if not title:
+ self.html_preprocess_sections = self.html_preprocess_sections + 1
+ self.log("found " + str(self.html_preprocess_sections) + " chapters. - " + str(chap))
+ return '<h2>'+chap+'</h2>\n'
+ else:
+ self.html_preprocess_sections = self.html_preprocess_sections + 1
+ self.log("found " + str(self.html_preprocess_sections) + " chapters & titles. - " + str(chap) + ", " + str(title))
+ return '<h2>'+chap+'</h2>\n<h3>'+title+'</h3>\n'
+
+ def chapter_break(self, match):
+ chap = match.group('section')
+ styles = match.group('styles')
+ self.html_preprocess_sections = self.html_preprocess_sections + 1
+ self.log("marked " + str(self.html_preprocess_sections) + " section markers based on punctuation. - " + str(chap))
+ return '<'+styles+' style="page-break-before:always">'+chap
+
+ def insert_indent(self, match):
+ pstyle = match.group('formatting')
+ span = match.group('span')
+ self.found_indents = self.found_indents + 1
+ if pstyle:
+ if not span:
+ return '<p '+pstyle+' style="text-indent:3%">'
+ else:
+ return '<p '+pstyle+' style="text-indent:3%">'+span
+ else:
+ if not span:
+ return '<p style="text-indent:3%">'
+ else:
+ return '<p style="text-indent:3%">'+span
+
+ def no_markup(self, raw, percent):
+ '''
+ Detects total marked up line endings in the file. raw is the text to
+ inspect. Percent is the minimum percent of line endings which should
+ be marked up to return true.
+ '''
+ htm_end_ere = re.compile('</p>', re.DOTALL)
+ line_end_ere = re.compile('(\n|\r|\r\n)', re.DOTALL)
+ htm_end = htm_end_ere.findall(raw)
+ line_end = line_end_ere.findall(raw)
+ tot_htm_ends = len(htm_end)
+ tot_ln_fds = len(line_end)
+ self.log("There are " + str(tot_ln_fds) + " total Line feeds, and " + str(tot_htm_ends) + " marked up endings")
+
+ if percent > 1:
+ percent = 1
+ if percent < 0:
+ percent = 0
+
+ min_lns = tot_ln_fds * percent
+ self.log("There must be fewer than " + str(min_lns) + " unmarked lines to add markup")
+ if min_lns > tot_htm_ends:
+ return True
+
+ def __call__(self, html):
+ self.log("********* Preprocessing HTML *********")
+ # Replace series of non-breaking spaces with text-indent
+ txtindent = re.compile(ur'<p(?P<formatting>[^>]*)>\s*(?P<span>(<span[^>]*>\s*)+)?\s*(\u00a0){2,}', re.IGNORECASE)
+ html = txtindent.sub(self.insert_indent, html)
+ if self.found_indents > 1:
+ self.log("replaced "+str(self.found_indents)+ " nbsp indents with inline styles")
+ # remove remaining non-breaking spaces
+ html = re.sub(ur'\u00a0', ' ', html)
+ # Get rid of empty <o:p> tags to simplify other processing
+ html = re.sub(ur'\s*<o:p>\s*</o:p>', ' ', html)
+ # Get rid of empty span tags
+ html = re.sub(r"\s*<span[^>]*>\s*</span>", " ", html)
+
+ # If more than 40% of the lines are empty paragraphs then delete them to clean up spacing
+ linereg = re.compile('(?<=<p).*?(?=</p>)', re.IGNORECASE|re.DOTALL)
+ blankreg = re.compile(r'\s*<p[^>]*>\s*(<(b|i|u)>)?\s*(</(b|i|u)>)?\s*</p>', re.IGNORECASE)
+ blanklines = blankreg.findall(html)
+ lines = linereg.findall(html)
+ if len(lines) > 1:
+ self.log("There are " + str(len(blanklines)) + " blank lines. " + str(float(len(blanklines)) / float(len(lines))) + " percent blank")
+ if float(len(blanklines)) / float(len(lines)) > 0.40:
+ self.log("deleting blank lines")
+ html = blankreg.sub('', html)
+ # Arrange line feeds and </p> tags so the line_length and no_markup functions work correctly
+ html = re.sub(r"\s*</p>", "</p>\n", html)
+ html = re.sub(r"\s*<p>\s*", "\n<p>", html)
+
+ # some lit files don't have any <p> tags or equivalent (generally just plain text between
+ # <pre> tags), check and mark up line endings if required before proceeding
+ if self.no_markup(html, 0.1):
+ self.log("not enough paragraph markers, adding now")
+ add_markup = re.compile('(?<!>)(\n)')
+ html = add_markup.sub('</p>\n<p>', html)
+
+ # detect chapters/sections to match xpath or splitting logic
+ heading = re.compile('<h[1-3][^>]*>', re.IGNORECASE)
+ self.html_preprocess_sections = len(heading.findall(html))
+ self.log("found " + str(self.html_preprocess_sections) + " pre-existing headings")
+ #
+ # Start with most typical chapter headings, get more aggressive until one works
+ if self.html_preprocess_sections < 10:
+ chapdetect = re.compile(r'(?=</?(br|p))(<(/?br|p)[^>]*>)\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<chap>(<[ibu]>){0,2}s*(<span[^>]*>)?\s*.?(Introduction|Synopsis|Acknowledgements|Chapter|Epilogue|Volume|Prologue|Book\s|Part\s|Dedication)\s*([\d\w-]+\:?\s*){0,8}\s*(</[ibu]>){0,2})\s*(</span>)?s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(p|/?br)>)\s*(<(/?br|p)[^>]*>\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<title>(<[ibu]>){0,2}(\s*[\w\'\"-]+){1,5}\s*(</[ibu]>){0,2})\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(br|p)>))?', re.IGNORECASE)
+ html = chapdetect.sub(self.chapter_head, html)
+ if self.html_preprocess_sections < 10:
+ self.log("not enough chapters, only " + str(self.html_preprocess_sections) + ", trying numeric chapters")
+ chapdetect2 = re.compile(r'(?=</?(br|p))(<(/?br|p)[^>]*>)\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<chap>(<[ibu]>){0,2}\s*.?(\d+\.?|(CHAPTER\s*([\dA-Z\-\'\"\?\.!#,]+\s*){1,10}))\s*(</[ibu]>){0,2})\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(p|/?br)>)\s*(<(/?br|p)[^>]*>\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<title>(<[ibu]>){0,2}(\s*[\w\'\"-]+){1,5}\s*(</[ibu]>){0,2})\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(br|p)>))?', re.UNICODE)
+ html = chapdetect2.sub(self.chapter_head, html)
+
+ if self.html_preprocess_sections < 10:
+ self.log("not enough chapters, only " + str(self.html_preprocess_sections) + ", trying with uppercase words")
+ chapdetect2 = re.compile(r'(?=</?(br|p))(<(/?br|p)[^>]*>)\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<chap>(<[ibu]>){0,2}\s*.?(([A-Z#-]+\s*){1,9})\s*(</[ibu]>){0,2})\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(p|/?br)>)\s*(<(/?br|p)[^>]*>\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(?P<title>(<[ibu]>){0,2}(\s*[\w\'\"-]+){1,5}\s*(</[ibu]>){0,2})\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</(br|p)>))?', re.UNICODE)
+ html = chapdetect2.sub(self.chapter_head, html)
+
+ # Unwrap lines
+ #
+ self.log("Unwrapping Lines")
+ # Some OCR sourced files have line breaks in the html using a combination of span & p tags
+ # span are used for hard line breaks, p for new paragraphs. Determine which is used so
+ # that lines can be un-wrapped across page boundaries
+ paras_reg = re.compile('<p[^>]*>', re.IGNORECASE)
+ spans_reg = re.compile('<span[^>]*>', re.IGNORECASE)
+ paras = len(paras_reg.findall(html))
+ spans = len(spans_reg.findall(html))
+ if spans > 1:
+ if float(paras) / float(spans) < 0.75:
+ format = 'spanned_html'
+ else:
+ format = 'html'
+ else:
+ format = 'html'
+
+ # Calculate Length
+ length = line_length(format, html, 0.4)
+ self.log("*** Median line length is " + str(length) + ",calculated with " + format + " format ***")
+ #
+ # Unwrap and/or delete soft-hyphens, hyphens
+ html = re.sub(u'\s*(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*', '', html)
+ html = re.sub(u'(?<=[-–—])\s*(?=<)(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?=[[a-z\d])', '', html)
+
+ # Unwrap lines using punctation if the median length of all lines is less than 200
+ unwrap = re.compile(r"(?<=.{%i}[a-z,;:\IA])\s*</(span|p|div)>\s*(</(p|span|div)>)?\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*" % length, re.UNICODE)
+ html = unwrap.sub(' ', html)
+
+ # If still no sections after unwrapping mark split points on lines with no punctuation
+ if self.html_preprocess_sections < 10:
+ self.log("Looking for more split points based on punctuation, currently have " + str(self.html_preprocess_sections))
+ #self.log(html)
+ chapdetect3 = re.compile(r'<(?P<styles>(p|div)[^>]*)>\s*(?P<section>(<span[^>]*>)?\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*(<[ibu]>){0,2}\s*(<span[^>]*>)?\s*.?([a-z#-*]+\s*){1,5}\s*\s*(</span>)?(</[ibu]>){0,2}\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</span>)?\s*</(p|div)>)', re.IGNORECASE)
+ html = chapdetect3.sub(self.chapter_break, html)
+ # search for places where a first or second level heading is immediately followed by another
+ # top level heading. demote the second heading to h3 to prevent splitting between chapter
+ # headings and titles, images, etc
+ doubleheading = re.compile(r'(?P<firsthead><h(1|2)[^>]*>.+?</h(1|2)>\s*(<(?!h\d)[^>]*>\s*)*)<h(1|2)(?P<secondhead>[^>]*>.+?)</h(1|2)>', re.IGNORECASE)
+ html = doubleheading.sub('\g<firsthead>'+'<h3'+'\g<secondhead>'+'</h3>', html)
+
+ return html
diff --git a/src/calibre/ebooks/fb2/output.py b/src/calibre/ebooks/fb2/output.py
index d0125afe89..d6c7a25a90 100644
--- a/src/calibre/ebooks/fb2/output.py
+++ b/src/calibre/ebooks/fb2/output.py
@@ -28,6 +28,9 @@ class FB2Output(OutputFormatPlugin):
])
def convert(self, oeb_book, output_path, input_plugin, opts, log):
+ from calibre.ebooks.oeb.transforms.jacket import linearize_jacket
+ linearize_jacket(oeb_book)
+
fb2mlizer = FB2MLizer(log)
fb2_content = fb2mlizer.extract_content(oeb_book, opts)
diff --git a/src/calibre/ebooks/html/input.py b/src/calibre/ebooks/html/input.py
index d57bfddd3e..084d48e54b 100644
--- a/src/calibre/ebooks/html/input.py
+++ b/src/calibre/ebooks/html/input.py
@@ -24,7 +24,7 @@ from calibre.constants import islinux, isfreebsd, iswindows
from calibre import unicode_path
from calibre.utils.localization import get_lang
from calibre.utils.filenames import ascii_filename
-from calibre.ebooks.conversion.preprocess import line_length
+from calibre.ebooks.conversion.utils import PreProcessor
class Link(object):
'''
@@ -491,20 +491,6 @@ class HTMLInput(InputFormatPlugin):
return (None, raw)
def preprocess_html(self, html):
- if not hasattr(self, 'log'):
- from calibre.utils.logging import default_log
- self.log = default_log
- self.log("********* Preprocessing HTML *********")
- # Detect Chapters to match the xpath in the GUI
- chapdetect = re.compile(r'(?=</?(br|p|span))(</?(br|p|span)[^>]*>)?\s*(?P<chap>(<(i|b)><(i|b)>|<(i|b)>)?(.?Chapter|Epilogue|Prologue|Book|Part|Dedication)\s*([\d\w-]+(\s\w+)?)?(</(i|b)></(i|b)>|</(i|b)>)?)(</?(p|br|span)[^>]*>)', re.IGNORECASE)
- html = chapdetect.sub('<h2>'+'\g<chap>'+'</h2>\n', html)
- # Unwrap lines using punctation if the median length of all lines is less than 150
- #
- # Insert extra line feeds so the line length regex functions properly
- html = re.sub(r"</p>", "</p>\n", html)
- length = line_length('html', html, 0.4)
- self.log.debug("*** Median length is " + str(length) + " ***")
- unwrap = re.compile(r"(?<=.{%i}[a-z,;:\IA])\s*</(span|p|div)>\s*(</(p|span|div)>)?\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*" % length, re.UNICODE)
- if length < 150:
- html = unwrap.sub(' ', html)
- return html
+ preprocessor = PreProcessor(log=getattr(self, 'log', None))
+ return preprocessor(html)
+
diff --git a/src/calibre/ebooks/lit/input.py b/src/calibre/ebooks/lit/input.py
index 9bf20fb1d4..65f5c607a2 100644
--- a/src/calibre/ebooks/lit/input.py
+++ b/src/calibre/ebooks/lit/input.py
@@ -6,10 +6,9 @@ __license__ = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
-import re
-
from calibre.customize.conversion import InputFormatPlugin
-from calibre.ebooks.conversion.preprocess import line_length
+from calibre.ebooks.conversion.utils import PreProcessor
+
class LITInput(InputFormatPlugin):
@@ -55,18 +54,6 @@ class LITInput(InputFormatPlugin):
def preprocess_html(self, html):
- self.log("********* Preprocessing HTML *********")
- # Detect Chapters to match the xpath in the GUI
- chapdetect = re.compile(r'(?=</?(br|p|span))(</?(br|p|span)[^>]*>)?\s*(?P<chap>(<(i|b)><(i|b)>|<(i|b)>)?(.?Chapter|Epilogue|Prologue|Book|Part|Dedication)\s*([\d\w-]+(\s\w+)?)?(</(i|b)></(i|b)>|</(i|b)>)?)(</?(p|br|span)[^>]*>)', re.IGNORECASE)
- html = chapdetect.sub('<h2>'+'\g<chap>'+'</h2>\n', html)
- # Unwrap lines using punctation if the median length of all lines is less than 150
- #
- # Insert extra line feeds so the line length regex functions properly
- html = re.sub(r"</p>", "</p>\n", html)
- length = line_length('html', html, 0.4)
- self.log("*** Median length is " + str(length) + " ***")
- unwrap = re.compile(r"(?<=.{%i}[a-z,;:\IA])\s*</(span|p|div)>\s*(</(p|span|div)>)?\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*" % length, re.UNICODE)
- if length < 150:
- html = unwrap.sub(' ', html)
- return html
+ preprocessor = PreProcessor(log=getattr(self, 'log', None))
+ return preprocessor(html)
diff --git a/src/calibre/ebooks/mobi/input.py b/src/calibre/ebooks/mobi/input.py
index 487e70c04f..b8dc7a9560 100644
--- a/src/calibre/ebooks/mobi/input.py
+++ b/src/calibre/ebooks/mobi/input.py
@@ -3,6 +3,7 @@ __license__ = 'GPL 3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
+import re
from calibre.customize.conversion import InputFormatPlugin
class MOBIInput(InputFormatPlugin):
@@ -37,3 +38,12 @@ class MOBIInput(InputFormatPlugin):
include_meta_content_type=False))
accelerators['pagebreaks'] = '//h:div[@class="mbp_pagebreak"]'
return mr.created_opf_path
+
+ def preprocess_html(self, html):
+ # search for places where a first or second level heading is immediately followed by another
+ # top level heading. demote the second heading to h3 to prevent splitting between chapter
+ # headings and titles, images, etc
+ doubleheading = re.compile(r'(?P<firsthead><h(1|2)[^>]*>.+?</h(1|2)>\s*(<(?!h\d)[^>]*>\s*)*)<h(1|2)(?P<secondhead>[^>]*>.+?)</h(1|2)>', re.IGNORECASE)
+ html = doubleheading.sub('\g<firsthead>'+'<h3'+'\g<secondhead>'+'</h3>', html)
+ return html
+
diff --git a/src/calibre/ebooks/oeb/transforms/cover.py b/src/calibre/ebooks/oeb/transforms/cover.py
index 59b42df68a..532c9bbc03 100644
--- a/src/calibre/ebooks/oeb/transforms/cover.py
+++ b/src/calibre/ebooks/oeb/transforms/cover.py
@@ -99,7 +99,8 @@ class CoverManager(object):
series_string = None
if m.series and m.series_index:
series_string = _('Book %s of %s')%(
- fmt_sidx(m.series_index[0], use_roman=True), m.series[0])
+ fmt_sidx(m.series_index[0], use_roman=True),
+ unicode(m.series[0]))
try:
from calibre.ebooks import calibre_cover
diff --git a/src/calibre/ebooks/oeb/transforms/flatcss.py b/src/calibre/ebooks/oeb/transforms/flatcss.py
index f48bdb9934..7212bd33c6 100644
--- a/src/calibre/ebooks/oeb/transforms/flatcss.py
+++ b/src/calibre/ebooks/oeb/transforms/flatcss.py
@@ -138,6 +138,7 @@ class CSSFlattener(object):
float(self.context.margin_left))
bs.append('margin-right : %fpt'%\
float(self.context.margin_right))
+ bs.extend(['padding-left: 0pt', 'padding-right: 0pt'])
if self.context.change_justification != 'original':
bs.append('text-align: '+ self.context.change_justification)
body.set('style', '; '.join(bs))
@@ -146,7 +147,6 @@ class CSSFlattener(object):
extra_css=css)
self.stylizers[item] = stylizer
-
def baseline_node(self, node, stylizer, sizes, csize):
csize = stylizer.style(node)['font-size']
if node.text:
@@ -194,7 +194,7 @@ class CSSFlattener(object):
value = 0.0
cssdict[property] = "%0.5fem" % (value / fsize)
- def flatten_node(self, node, stylizer, names, styles, psize, left=0):
+ def flatten_node(self, node, stylizer, names, styles, psize, item_id, left=0):
if not isinstance(node.tag, basestring) \
or namespace(node.tag) != XHTML_NS:
return
@@ -286,15 +286,18 @@ class CSSFlattener(object):
if self.lineh and 'line-height' not in cssdict:
lineh = self.lineh / psize
cssdict['line-height'] = "%0.5fem" % lineh
+
if (self.context.remove_paragraph_spacing or
self.context.insert_blank_line) and tag in ('p', 'div'):
- for prop in ('margin', 'padding', 'border'):
- for edge in ('top', 'bottom'):
- cssdict['%s-%s'%(prop, edge)] = '0pt'
+ if item_id != 'calibre_jacket' or self.context.output_profile.name == 'Kindle':
+ for prop in ('margin', 'padding', 'border'):
+ for edge in ('top', 'bottom'):
+ cssdict['%s-%s'%(prop, edge)] = '0pt'
if self.context.insert_blank_line:
cssdict['margin-top'] = cssdict['margin-bottom'] = '0.5em'
if self.context.remove_paragraph_spacing:
cssdict['text-indent'] = "%1.1fem" % self.context.remove_paragraph_spacing_indent_size
+
if cssdict:
items = cssdict.items()
items.sort()
@@ -313,7 +316,7 @@ class CSSFlattener(object):
if 'style' in node.attrib:
del node.attrib['style']
for child in node:
- self.flatten_node(child, stylizer, names, styles, psize, left)
+ self.flatten_node(child, stylizer, names, styles, psize, item_id, left)
def flatten_head(self, item, stylizer, href):
html = item.data
@@ -360,7 +363,7 @@ class CSSFlattener(object):
stylizer = self.stylizers[item]
body = html.find(XHTML('body'))
fsize = self.context.dest.fbase
- self.flatten_node(body, stylizer, names, styles, fsize)
+ self.flatten_node(body, stylizer, names, styles, fsize, item.id)
items = [(key, val) for (val, key) in styles.items()]
items.sort()
css = ''.join(".%s {\n%s;\n}\n\n" % (key, val) for key, val in items)
diff --git a/src/calibre/ebooks/oeb/transforms/jacket.py b/src/calibre/ebooks/oeb/transforms/jacket.py
index fec4d230c3..88c7a4ff0e 100644
--- a/src/calibre/ebooks/oeb/transforms/jacket.py
+++ b/src/calibre/ebooks/oeb/transforms/jacket.py
@@ -6,139 +6,200 @@ __license__ = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
-import textwrap
+import sys
from xml.sax.saxutils import escape
-from itertools import repeat
from lxml import etree
-from calibre.ebooks.oeb.base import XPath, XPNSMAP
-from calibre import guess_type
+from calibre import guess_type, strftime
+from calibre.ebooks.BeautifulSoup import BeautifulSoup
+from calibre.ebooks.oeb.base import XPath, XHTML_NS, XHTML
from calibre.library.comments import comments_to_html
+
+JACKET_XPATH = '//h:meta[@name="calibre-content" and @content="jacket"]'
+
class Jacket(object):
'''
Book jacket manipulation. Remove first image and insert comments at start of
book.
'''
- JACKET_TEMPLATE = textwrap.dedent(u'''\
- <html xmlns="%(xmlns)s">
- <head>
- <title>%(title)s
-
-
-
-
-
-
%(title)s
-
%(jacket)s
-
%(series)s
-
%(rating)s
-
%(tags)s
-
-
- %(comments)s
-
-